We are studying the molecular mechanisms controlling the segregation (delamination) of neuronal precursors in Drosophila. In order to move out of the neurectodermal epithelium, neuronal precursors undergo active changes of their cytoskeleton and contacts to neighboring cells. Since, according to recent findings, many of the genes controlling these processes are involved in the delamination of other (non-neural) tissues as well, our studies have a bearing on morphogenetic movements in general. These studies ultimately are also of clinical importance, because the molecules controlling cell movements during normal development (e.g., cadherins; homologs of Drosophila Notch and wingless genes) also play a central role in neoplastic growth. The specific objectives of this proposal are the molecular analysis of two genes involved in neuronal precursor segregation, faint sausage (fas) and shotgun (shg). Mutations in fas cause cytoskeletal changes in many epithelial cells, associated with the loss of polarity and monolayered arrangement of these cells. In the neurectoderm, these changes lead to defects in neuronal precursor delamination. In shg mutant embryos, there is a widespread degeneration of the neurectoderm (as well of other specialized epithelia). In the foregoing granting period, both fas and shg were characterized phenotypically and genetically. We have cloned fas with the help of a P1 plasmid crossing the fas breakpoint and a PlacZ insertion in fas. Cloning of shg was initiated by starting a walk in a cosmid library from a P1 plasmid. In this application experiments are proposed to identify the fas and shg transcripts and characterize their sequence and expression pattern. For both fas and shg, fusion proteins expressed in E. coli will be used to generate antibodies. Germline transformation of our isolated genomic clones will be attempted to rescue the mutations. We will screen for additional mutations affecting early neurogenesis. Work of the foregoing granting period had established that there exist similarities between cytoskeletal changes in delamination and mitosis, and that the pattern of these two processes in the neurectoderm are closely correlated. We have shown that two genes known to control elamination, Notch and wingless, have an effect on mitosis of neurectodermal cells as well. We here propose a set of developmental- genetic experiments which address the relationship between delamination and mitosis, and their control by neurogenic and segment polarity genes, in greater detail.